This is going to sound harsh. Your idea is exciting, and I don't want to discourage experimentation, but I don't think it's possible. To be clear, I do want to discourage experimentation when it puts the welfare of animals at risk.
What exactly do you mean by "self-sustainable"? Do you just mean in terms of filtration needs, or do you mean a closed loops system that does not need food added to it? For the size food chain you propose, to not add any food at all you would almost certainly require something pond-sized.
If you are planning on making a no-food-input tank, with a food chain of algae<shrimp/snails<small fish<predator fish, you will not succeed. The fish and shrimp that you are thinking can live for a long time on algae need, at some point, a source of protein. In the wild, this would typically be small copepods, but even very large aquariums will be stripped of copepod colonies with a couple of fish in them. The higher predators that eat smaller fish will, if fed nothing else, eliminate the small fish population in an aquarium.
In a fairly large tank, I think you could pull this off with dwarf shrimp, but I would still be worried about malnutrition. Perhaps you could throw a small dead fish in every once in awhile to simulate a scavenger ecosystem. Even Opae ula shrimp can't live on algae forever - those "self sustaining" kits you see for sale are just mini starvation chambers.